lkml.org 
[lkml]   [2010]   [Sep]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC][PATCH] big continuous memory allocator v2
On Tue, 07 Sep 2010 09:29:21 +0200
Andi Kleen <andi@firstfloor.org> wrote:

> KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> writes:
>
> > This is a page allcoator based on memory migration/hotplug code.
> > passed some small tests, and maybe easier to read than previous one.
>
> Maybe I'm missing context here, but what is the use case for this?
>

I hear some drivers want to allocate xxMB of continuous area.(camera?)
Maybe embeded guys can answer the question.

> If this works well enough the 1GB page code for x86, which currently
> only supports allocating at boot time due to the MAX_ORDER problem,
> could be moved over to runtime allocation. This would make
> GB pages a lot nicer to use.
>
> I think it would still need declaring a large moveable
> area at boot right? (but moveable area is better than
> prereserved memory)
>
Right.

I think a main use-case is using allocation-at-init rather than boot
option. If modules can allocate a big chunk in __init_module() at boot,
boot option will not be necessary and it will be user friendly.
I think there are big free space before application starts running.

If on-demand loading of modules are required, it's safe to use MOVABLE zones.

> On the other hand I'm not sure the VM is really up to speed
> in managing such large areas.
>

Thanks,
-Kame



\
 
 \ /
  Last update: 2010-09-07 10:33    [from the cache]
©2003-2011 Jasper Spaans